Skip to content

TriggerMesh AMI

This document walks you through deploying the TriggerMesh AMI (beta) on the Amazon AWS platform.

Warning

This installation method is meant to provide a fast mechanism to test TriggerMesh on AWS. It is not meant to be used in a production environment.

Pre-requisites

Creating the TriggerMesh EC2 Instance

Log in to the Amazon AWS console and create a new EC2 instance in the us-west-1 region.

Search for the TriggerMesh AMI in Community AMIs and select it for the EC2 instance.

TriggerMesh AMI Search

The TriggerMesh AMI runs the TriggerMesh Cloud Native Integration Platform (and its dependencies) on top of a single node Kubernetes cluster using (k3s). The recommended instance type for the EC2 instance is t2.2xlarge or higher.

The instance requires 32GiB or higher as required.

For the Knative services to be accessible externally over the HTTP protocol, configure the EC2 instance security group to allow HTTP traffic to the instance as shown in the screenshot below:

EC2 Security Groups

Launch the EC2 instance after specifying the SSH keypair for logging into the instance.

Logging into the instance

As the user rancher, SSH into the EC2 instance using its public IPv4 address.

EC2 Instance Details

Verifying the deployment

After logging into the EC2 instance, verify that the triggermesh-controller and the triggermesh-webhook deployments are running successfully in the triggermesh namespace.

$ kubectl get deploy -n triggermesh
NAME                     READY   UP-TO-DATE   AVAILABLE   AGE
triggermesh-controller   1/1     1            1           29m
triggermesh-webhook      1/1     1            1           29m

To check if Knative services are accessible externally, deploy Sockeye to the cluster.

$ kubectl apply -f https://github.com/n3wscott/sockeye/releases/download/v0.7.0/release.yaml
service.serving.knative.dev/sockeye configured

Now try to access the sockeye Knative service you just deployed using the URL listed by the following command:

$ kubectl get ksvc sockeye
NAME      URL                                             LATESTCREATED   LATESTREADY     READY   REASON
sockeye   http://sockeye.default.54.219.32.xxx.sslip.io   sockeye-00001   sockeye-00001   True

If the deployment was successful, the Sockeye webpage should load without any errors.

You may now delete the sockeye Knative service:

$ kubectl delete ksvc sockeye
service.serving.knative.dev "sockeye" deleted

Next steps